ABSTRACT
The idea behind process synthesis and design is the selection of the topology, the šowsheet, and the operating conditions to transform a set of raw materials into products [1]. For a long time, the objective was to design the process for the minimum cost (or maximum pro’t). However, environmental concerns have transformed this problem into a multiobjective one so that we account not only for economic issues but also for environmental ones such as greenhouse gas emissions or freshwater consumption, whose impact cannot be easily addressed via economic evaluations alone. We can distinguish two major methods: the use of heuristics and physical know-how, such as hierarchical decomposition [2] and pinch analysis [3], or the use of superstructure optimization [4,5]. The more recent approach consists of combining both, where we ’rst develop a superstructure of alternative technologies for the various steps of the transformation of raw materials to products [6]; we model the different technologies using different approaches such as ’rst principles, mass and energy balances, chemical equilibria, or surrogate models based on rules of thumb, experimental correlations, design of experiments, or reactor kinetics [7] formulating a problem whose solution yields the optimal production
11.1 Introduction .................................................................................................. 481 11.2 Hints for Modeling Processes in GAMS® .................................................... 482 11.3 Examples ....................................................................................................... 483
11.3.1Process Operation ............................................................................. 483 11.3.1.1Ammonia Reactor .............................................................. 483 11.3.1.2SO2 Catalytic Converter ..................................................... 489 11.3.1.3 Steam Reforming of Natural Gas ...................................... 493
11.3.2Technology Selection: Process Design ............................................. 501 11.3.3Further Reading: Simultaneous Optimization
and Topology Design........................................................................506 References ..............................................................................................................508
process. This type of formulation involves discrete and continued decisions for the selection of technologies and operating conditions, giving rise to a mixed-integer nonlinear programming problem (MINLP) or a generalized disjunctive programming (GDP) model.
